A rate-splitting approach to robust multiuser MISO transmission

For multiuser MISO systems with bounded uncertainties in the Channel State Information (CSI), we consider two classical robust design problems: maximizing the minimum rate subject to a transmit power constraint, and power minimization under a rate constraint. Contrary to conventional strategies, we propose a Rate-Splitting (RS) strategy where each message is divided into two parts, a common part and a private part. All common parts are packed into one super common message encoded using a shared codebook and decoded by all users, while private parts are independently encoded and retrieved by their corresponding users. We prove that RS-based designs achieve higher max-min Degrees of Freedom (DoF) compared to conventional designs (NoRS) for uncertainty regions that scale with SNR. For the special case of non-scaling uncertainty regions, RS contrasts with NoRS and achieves a non-saturating max-min rate. In the power minimization problem, RS is shown to combat the feasibility problem arising from multiuser interference in NoRS. A robust design of precoders for RS is proposed, and performance gains over NoRS are demonstrated through simulations.